low reynolds number models in Fluent

doug · April 16, 2009, 17:29

I have two questions:

1. Does the term "low Reynolds number model" mean that the model is only good for low Reynolds number flows? Or does it mean that it is a modification capable of handling both hi and low Reynolds number flows?

2. When I run fluent using periodic boundary conditions and either the Lam Bremhorst (index 1) or Launder Sharma (index 2) model at a Reynolds number of 10,000 the kinetic energy eventually goes to zero across the channel, and I get a laminar solution. Any ideas why that might be? This does not happen when I run the standard k-e model with standard wall functions or the standard k-e model with "enhanced wall treatment" with periodic boundary conditions.

Thanks.

Doug

harishg · April 16, 2009, 17:47

It means that your model can be integrated upto the wall and you do not need to use any wall function. Check the value of yplus. The wall function approach requires your first grid point to be in the log layer (yplus > 30) while the low Re models requires yplus of the first grid point to be less than 5 to accurately model the near wall effects.

doug · April 16, 2009, 18:19

My first wall plus value is around .01 on my current grid (100x100). I've run some more cases and find that if I coarsen the grid (50x50) I get the turbulent velocity profile as expected. However, if I refine the grid (200x200) I get the laminar solution. This is opposite of my intuition which would say that coarsening the grid would give the worse result...?

doug · April 16, 2009, 18:21

Wait a minute. I take that back. The 200x200 grid is giving me a better velocity profile. I'll get back when I have better results.

Thanks for your help!

doug · April 17, 2009, 16:43

Does anyone know what Fluent is using for the BCs for the low-re number models available only through the TUI? I've searched their documentation in vain to find the boundary conditions for these models. I can find them from the original literature from the models, but I'm a little wary that Fluent is fudging things near the wall...

Thanks!

neazen · April 20, 2009, 11:42

About low reynolds number flow question :

Usually this term refer to specific flows such that the Reynolds number is very small so the inertia term of the Navier-Stokes equation is very small compared to the viscous forces and it is possible to neglect it. (Like for lubrification problems)
This leads to the so called (linear) Stokes equations for a viscous fluid, were the Reynolds number can be replaced by unity.
Obviously in this case the model is inappropriate to flows with high Re number.

Now the question is : does Fluent mean something else by using this term?

asurpam · August 4, 2012, 15:39

hi,
Can anyone tell me how to set periodic boundary condition for low Reynolds number flow in a pipe???

April 16, 2009, 17:29	low reynolds number models in Fluent	#1
doug Member Doug Hunsaker Join Date: Mar 2009 Location: Logan, UT Posts: 63 Rep Power: 17	I have two questions: 1. Does the term "low Reynolds number model" mean that the model is only good for low Reynolds number flows? Or does it mean that it is a modification capable of handling both hi and low Reynolds number flows? 2. When I run fluent using periodic boundary conditions and either the Lam Bremhorst (index 1) or Launder Sharma (index 2) model at a Reynolds number of 10,000 the kinetic energy eventually goes to zero across the channel, and I get a laminar solution. Any ideas why that might be? This does not happen when I run the standard k-e model with standard wall functions or the standard k-e model with "enhanced wall treatment" with periodic boundary conditions. Thanks. Doug

April 20, 2009, 11:42		#6
neazen New Member Join Date: Apr 2009 Posts: 8 Rep Power: 17	About low reynolds number flow question : Usually this term refer to specific flows such that the Reynolds number is very small so the inertia term of the Navier-Stokes equation is very small compared to the viscous forces and it is possible to neglect it. (Like for lubrification problems) This leads to the so called (linear) Stokes equations for a viscous fluid, were the Reynolds number can be replaced by unity. Obviously in this case the model is inappropriate to flows with high Re number. Now the question is : does Fluent mean something else by using this term? __________________ If a problem can be solved, there is nothing to worry about. If it can't be solved, worrying will do no good. So be Zen and stay Zen

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Implementing low Re number k-e models in Fluent.	kiran	FLUENT	1	July 3, 2017 18:04
low reynolds number K-epsilon model	amar	FLUENT	25	October 7, 2015 15:07
[blockMesh] BlockMeshmergePatchPairs	hjasak	OpenFOAM Meshing & Mesh Conversion	11	August 15, 2008 08:36
Low Reynolds number flow	Terje	CFX	3	June 20, 2007 09:10
Low Reynolds Number Flow Computation	A.M.Yang	Main CFD Forum	7	January 22, 2002 13:01

April 16, 2009, 17:47		#2
harishg Senior Member N/A Join Date: Mar 2009 Posts: 189 Rep Power: 17	It means that your model can be integrated upto the wall and you do not need to use any wall function. Check the value of yplus. The wall function approach requires your first grid point to be in the log layer (yplus > 30) while the low Re models requires yplus of the first grid point to be less than 5 to accurately model the near wall effects.

April 16, 2009, 18:19		#3
doug Member Doug Hunsaker Join Date: Mar 2009 Location: Logan, UT Posts: 63 Rep Power: 17	My first wall plus value is around .01 on my current grid (100x100). I've run some more cases and find that if I coarsen the grid (50x50) I get the turbulent velocity profile as expected. However, if I refine the grid (200x200) I get the laminar solution. This is opposite of my intuition which would say that coarsening the grid would give the worse result...?

April 16, 2009, 18:21		#4
doug Member Doug Hunsaker Join Date: Mar 2009 Location: Logan, UT Posts: 63 Rep Power: 17	Wait a minute. I take that back. The 200x200 grid is giving me a better velocity profile. I'll get back when I have better results. Thanks for your help!

April 17, 2009, 16:43		#5
doug Member Doug Hunsaker Join Date: Mar 2009 Location: Logan, UT Posts: 63 Rep Power: 17	Does anyone know what Fluent is using for the BCs for the low-re number models available only through the TUI? I've searched their documentation in vain to find the boundary conditions for these models. I can find them from the original literature from the models, but I'm a little wary that Fluent is fudging things near the wall... Thanks!

August 4, 2012, 15:39		#7
asurpam New Member Ayushi Join Date: Jul 2012 Posts: 4 Rep Power: 14	hi, Can anyone tell me how to set periodic boundary condition for low Reynolds number flow in a pipe???